Court name
Court of Appeal of Tanzania

Salum Mhando vs Republic () [1993] TZCA 15 (28 July 1993);

Law report citations
1993 TLR 170 (TZCA)
Media neutral citation
[1993] TZCA 15

Mfalila, J.A., delivered the following considered judgment of the court:
I The appellant Salum Mhand was convicted of two counts of rob-

bery with violence contrary to s 285 and s 286 of the Penal Code in Criminal Case No 1589/89 at A Kivukoni District Court in which he was jointly charged with two other persons. Upon his conviction, the appellant who was the first accused at the trial, was sentenced to fifteen years imprisonment on each count but the sentences were ordered to run concurrently. Against this conviction and sentence B the appellant unsuccessfully appealed to the High Court. The High Court (Bahati J) held that the appellant had been sufficiently identified by the complainants, the victims of the robbery but that since the appellant and his associates were armed at the time of committing the offence, they were liable to be sentenced to imprisonment for not less than thirty years. Accordingly the learned Judge set C aside the sentence of fifteen years on each count and substituted a sentence of thirty years imprisonment. We assume that the learned Judge passed the sentence of thirty years imprisonment on each count and that the sentences were to run concurrently. But at the same time we think the D judge should have invoked the court's revisional jurisdiction to rectify the sentences in respect of the third accused. From this decision, the appellant further appealed to this Court arguing at the hearing that he was not properly identified as being part of the gang which broke into the shops of the E complainants, and in his memorandum of appeal, he complained that as a juvenile he was exempted from the rigours of the Minimum Sentence Act and that therefore he should not have been sentenced to the minimum sentence prescribed by the Minimum Sentences Act.
Before we deal with the appeal on merits, we wish to draw the attention of the Registrar of the High F Court to the unsatisfactory way in which some criminal records of appeal are being prepared. We would like to ask him to certify to us only truly correct records. The act of certifying as correct a record of appeal is not an automatic function. In the present case, the record contained a wrong G charge sheet which was withdrawn. The correct charge sheet with additional accused persons was not made part of the record, hence we had to resort to the original record to see whether the charges referred to in the evidence were the same as those on the new charge sheet and who were the H accused and their numbering. In future we may not take kindly to registrars who certify as correct incorrect records.
The main question before us as it was in the Trial Court and in the High Court was the identification of the appellant. The appellant contends that he was not a member of the gang which raided the two I shops on 27 September 1989. At the trial two witnesses gave

A evidence in support of the charge in Count 1. These are the complainant Sifuni Fadhili (PW1) and another occupant of the house Mikundiwe Zakaya (PW2). What was the substance and effect of the evidence of these two witnesses in support of the charge in Count 1. PW1 related how a group of B about twenty people broke into his shop and attacked the occupants with an assortment of weapons including a panga and iron bar. The witness was assaulted with a panga and another occupant with an iron bar before the assailants made away with a large quantity of articles from the shop. The C witness described what each of the three accused persons did and the articles he took with him. With regard to the appellant, the witness said that he knew him because he always went to his shop. All this is very well, but nowhere does this witness explain how he identified the appellant among the gang of twenty people in what could only have been a room without any light. The raid took place D between 1 am and 2 am. It was extremely important for this witness to explain how he identified the appellant in these circumstances. The other witness PW2 gave similar evidence describing what weapon each accused carried and whom he attacked. But like PW1 he did not explain how he identified the appellant in this darkness. This is how he explained the way he identified the accused. E He said:
`I identify both the accused persons were in the scene of a crime.' (sic)
F We think this hardly says anything regarding the identification of the appellant although later he states that he knew the appellant before the incident. When cross-examined by the appellant, the witness simply replied that he identified him because he knew him before. The question remains, how did he do this identification in darkness. Quite clearly then the charge in Count 1 was not proved G as the identification of the appellant to have been a member of the twenty strong gang was not established beyond reasonable doubt.
With regard to the charge in Count 2, the relevant witness was the complainant Tito Mbugi (PW4), H and another occupant in the room Mohamed Abdallah (PW6). PW4 described in his evidence how on 27 September 1989 at 1 am the shop was raided by a group of people wielding iron bars, axes and pangas. On entering the room, they shone their torches and threw a stone at him. He added that the gang was big and he could not identify the number. They left after robbing him of various shop I goods and cash. They then went to a neighbour's shop. He concluded his evidence by saying

that that night he identified all the accused. The same question as in Count 1 can be asked, namely A how did he manage to identify the accused that night. The witness provided the answer, he said:
`After breaking the door, they lighted the torches, they threw stone on me, and got in the room and lighted the light.' B
So according to this witness, the bandits put on the light when they entered the room although they had torches and actually had them on. Is this a reasonable scenario that thieves could deliberately expose themselves to clear identification - by putting on the light in the room. We have strong C reservations on the truthfulness of this statement. These reservations are strengthened by the evidence of the other witness PW6 who stated:
`It was 2 am. They pass through at the backyard they just pushed the door and was down, when the bandit arrived D we waked up stated that nobody should be out "Sisi tumefuata kazi moia tu stakayefika aakiona". At the backyard there is light but succeed to make light off - they broke the torches light the whole house was dark. They broke the shop and don't know the property which had been taken.'
Although the passage could be improved grammatically, the general meaning is clear that the E bandits put off the light and plunged the whole house in darkness. We think this version makes sense, that the bandits armed with torches, plunged the house in darkness to make their operations easier by using their torches while at the same time making their identification difficult. In the light of this evidence, we think the learned Judge on first appeal was unduly optimistic when he stated: F
`The witnesses were quite positive about the identity of the appellant. Despite any contradictions there may be in their evidence, the evidence on the identification is very clear. They knew the appellant before the day in question. G Although there was only torch light in the shop, it was in my view, enough to enable witnesses to identify the appellant who they knew very well.'
Like us the learned Judge also seems to have disbelieved or doubted PW4's assertion that the H bandits put on the light when they entered the building, hence his finding that there was only torch light in the room. But as the learned Judge would most certainly appreciate, torch light in a robbery at night, does not help the victim to identify the attackers, on the contrary it prevents him from I

A identifying them, while making it easier for the attackers to view the surroundings. It is quite obvious then that the appellant was arrested on suspicion because the complainant knew him to be virtually a `bhang' smoking vagabond.
B As it will be noticed, we have taken the unusual step in this appeal of interfering with concurrent findings of fact made by two courts below. On a second appeal to this Court, we are only supposed to deal with questions of law. But this approach rests on the premise that the findings of fact are based on a correct appreciation of the evidence. If as in this case both courts completely misapprehend the substance, nature and quality of the evidence, resulting in an unfair conviction, this C Court must in the interests of justice intervene. In the High Court the learned Judge approached the case globally forgetting that the appellant was charged with two counts which had to be considered separately. The consequences of this approach was that the learned Judge failed to analyse the D evidence touching on each count, this led him to make general statement which cannot be supported by the evidence led in support of each count. In a case in which a similar problem arose, DPP v Jaffari Mfaume Kawawa (1) this Court held as follows at 153:
E `The next important point for consideration and decision in this case is whether it is proper for this Court to evaluate the evidence afresh and come to its own conclusions in matters of fact. This is a second appeal brought under the provisions of s 5(7) of the Appellate Jurisdiction Act 1979. The appeal therefore lies to this Court only on a point or F points of law. Obviously this position applies only where there are no misdirections or non-directions on the evidence, a court of second appeal is entitled to look at the relevant evidence and make its own findings of fact.'
G We are therefore satisfied that the identification of the appellant was not put outside the pail of doubt to make his conviction safe. Accordingly we allow the appeal, quash the convictions on both counts and order that the appellant Salum Mhando be released from prison immediately unless he is H lawfully held on some other cause.